Mobile computing devices are ubiquitous in the modern world. People are constantly using their smartphones, tablet computers, and/or other portable devices. These devices can track when the user is using the device, the location where the user is using the device, environmental conditions around the device, and/or other indicia of use. The computing devices can analyze this usage data (e.g., indicia of use) and determine various behaviors or patterns of activity associated with the user of the device. For example, a user's pattern of use of a device (e.g., a smartphone) can be analyzed to predict the user's sleep period patterns and/or detect when a user is sleeping. However, if a user uses multiple computing devices, the sleep prediction and/or sleep detection performed by the user's primary computing device may be erroneous.